(3.237.97.64) 您好!臺灣時間:2021/03/03 00:31
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:林谷炘
研究生(外文):Ku-Sin Lin
論文名稱:長期匯率預測效率之研究:GRU神經網路之比較應用
論文名稱(外文):Forcasting Accuracy of Long-Term Exchange Rate using GRU Neural Network model
指導教授:李顯鋒李顯鋒引用關係
指導教授(外文):Hsien-Feng Lee
口試委員:謝德宗羅光達
口試日期:2019-06-05
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:經濟學研究所
學門:社會及行為科學學門
學類:經濟學類
論文種類:學術論文
論文出版年:2019
畢業學年度:107
語文別:中文
論文頁數:47
中文關鍵詞:匯率預測深度學習類神經網路遞迴式神經網路 (RNN)Gated Recurrent Unit(GRU)
DOI:10.6342/NTU201900914
相關次數:
  • 被引用被引用:0
  • 點閱點閱:83
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
匯率預測在經濟領域一直都是難解的問題,近幾十年不斷有新的理論問世,但實證結果卻仍 然難以打敗隨機漫步模型。而匯率資料係屬非線性,非定態及高雜訊的資料,使用非線性的 模型進行預測較為合理,故本文選擇了幾個匯率模型,並以 Gated Recurrent Unit(GRU) 神經 網路的方法測試之。
本文建立了一個兩層 GRU 神經網路,使用兩階段及單一方程式兩種預測方式,搭配兩 個匯率決定模型以及一個均衡匯率模型,測試 GRU 神經網路模型的表現是否優於隨機 漫步模型及最小平方法迴歸。實驗結果發現,在使用兩階段預測方式預測匯率走勢時, 各個期數 GRU 神經網路模型能夠表現得比最小平方法迴歸好,但除了五年後匯率的預 測外,其他期數表現都難以穩定勝過隨機漫步。不過在預測漲跌方向的變換上,預測 期間超過兩年時 GRU 神經網路模型都能有效贏過隨機漫步及最小平方法迴歸。在使用 單一方程式預測時可以則發現,無論是預測匯率走勢還是漲跌方向變換,在預測期間超 過半年時,GRU 神經網路模型普遍都優於隨機漫步模型,卻難以穩定勝過最小平方法迴歸。
Exchange rate forecasting has always been a puzzle in the field of Finance and Economics. Numerous models came out during the last few decades, but empirical works found that there were few evidence these models can defeat the random walk model.On the other hand, exchange rate data is typically nonlinear, non-stationary and contains high noise. Thus it is appropriate to use a nonlinear approach to do prediction.
In this paper, we build a two layer Gated Recurrent Unit(GRU) recurrent network model, cooper- ate with a two-step and a single-equation method, and test on two exchange decision model and an equilibrium exchange rate model to see whether the deep learning model can outperform the random walk model and OLS regression model. We discover that, with the two-step method, when comparing MSE, deep learning model can outperform OLS in any period. But except when k = 60, deep learning cannot outperform the random walk model. When it comes to predict the direct of change, deep learn- ing model can outperform random walk model and OLS model when k ≥ 6. With single-equation method, both in MSE and direction of change, deep learning model can outperform the random walk model but struggles to defeat OLS model.
口試委員審定書 i
誌謝 ii
中文摘要 iii
Abstract iv
1 緒論 1
2 匯率理論介紹 3
2.1 價格僵固貨幣模型 3
2.1.1 購買力平價理論 3
2.1.2 位拋補利率評價理論 4
2.1.3 價格僵固貨幣模型 4
2.2 行為均衡匯率模型 5
3 演算法介紹 8
3.1 類神經網路 8
3.2 時間遞迴神經網路 9
3.3 GRU 11
3.4 放棄法 13
4 實驗設計 14
4.1 資料描述 14
4.2 模型設置 15
4.3 模型訓練方式 16
4.4 預測效果衡量 17
4.4.1 比較基準 17
4.4.2 比較方式 18
5 實驗結果 20
6 結論與未來展望 25
References 26
A 附錄 29
A.1 倒傳遞法 29
A.2 梯度消失問題 30
A.3 LSTM結構 32
A.4 第一隱藏層GRU單元個數 34
A.5 實驗數據 36
[1] Akita, R., Yoshihara, A., Matsubara, T., Uehara, K. (2016). ”Deep Learning for Stock Predic- tion Using Numerical and Textual Information”, IEEE/ACIS 15th International Conference on Computer and Information Science (ICIS), 1-6.
[2] Athiwaratkun, B., and Stokes, J.W. (2017). ”Malware Classification with LSTM and GRU Lan- guage Models and a Character-Level CNN”, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2482-2486.
[3] Balassa,B.(1964)“. ThePurchasingPowerParityDoctrine:AReappraisal,”JournalofPolitical Economy 72 , 584-96.
[4] Clark, P. B. and MacDonald, R. (1998). “Exchange Rates and Economic Fundamentals: A Methodological Comparison of Beers and Feers,”Equilibrium Exchange Rates, 285-322. Dor- drecht: Springer Netherlands. doi: 10.1007/978-94-011-4411-7_10
[5] Clark, P. B. and MacDonald, R. (2004). “Filtering the BEER: A Permanent and Transitory Decomposition,”Global Finance Journal 15 (1), 29-56. doi: 10.1016/j.gfj.2003.10.005
[6] Chen, S-S. and Chou,Y-H. (2010). ”Exchange Rates and Fundamentals: Evidence from Long- Horizon Regression Tests,” Oxford Bulletin of Economics and Statistic, 72, 1, 63-88. doi: 10.1111/j.1468-0084.2009.00571.x
[7] Cho, K., Bahdanau ,D., Bougares, F., Schwenk, H. and Bengio, Y. (2014). ”Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation”. arXiv: 1406.1078 [cs.CL].
[8] Cheung, Y-W, Chinn, M. D., Pascual, A.D. and Zhang, Y. (2017). ”Exchange Rate Prediction Redux: New Models, New Data, New Currencies,” European Central Bank Working Paper, No 2018. doi: 10.2866/524144
[9] Diebold,F.X.andMariano,R.S.(1995).”ComparingPredictiveAccuracy”,JournalofBusiness & Economic Statistics, 20:1, 134-144. doi: 10.1198/073500102753410444
[10] Dornbusch, R.(1976). ”Expectations and Exchange Rate Dynamics,” Journal of Political Econ- omy, Vol. 84, No. 6, 1161-1176.
[11] Fu, R., Zhang, Z. and Li, L. (2016) .”Using LSTM and GRU Neural Network Methods for Traffic Flow Prediction,” Proc. 4th Conf. Chin. Assoc. Autom. Acad. Annu. (YAC), 324-328. doi: 10.1109/YAC.2016.7804912
[12] Galeshchuk,S.andMukherjee,S.(2017).”DeepNetworksforPredictingDirectionofChangein Foreign Exchange Rates,” Intelligent Systems in Accounting Finance & Management,100-110. doi: 10.1002/isaf.1404
[13] Irie, K., Tüske, Z., Alkhouli, T., Schlüter, R. and Ney, H.(2016). ”LSTM, GRU, Highway and a Bit of Attention:An Empirical Overview for Language Modeling in Speech Recognition,” In- terspeech, San Francisco, CA, USA, 2016. doi: 10.21437/Interspeech.2016-491
[14] Kuan,C-M.andLiu,T.(1995).”ForecastingExchangeRatesUsingFeedforwardandRecurrent Neural Networks,” Journal of Applied Econometrics vol 10, 347-364.
[15] Hochreiter, S. Bengio, Y. Frasconi, P. and Schmidhuber, J. (2001). ”Gradient flow in recurrent nets: the difficulty of learning long-term dependencies.” A Field Guide to Dynamical Recurrent Neural Networks. New York: IEEE Press.
[16] Hochreiter, S. and Schmidhuber, J. (1997). ”Long Short-Term Memory,” Neural computation, 9(8),1735–1780.
[17] Meese, R. and Rogoff, K. (1983). ”Empirical Exchange Rate Models of the Seventies: Do They Fit Out of Sample?” Journal of International Economics 14,3-24. (a)
[18] Nelson, D. M. Q., Pereira, A. C. M. and R. A. de Oliviera. (2017). ”Stock Market’s Price Move- ment Prediction with LSTM Neural Networks”, International Joint Conference on Neural Net- works (IJCNN), pp. 1419-1426. doi: 10.1109/IJCNN.2017.7966019
[19] Rumelhart, D. E., Hinton, G. E. and Williams, R. J. (1986). ”Learning representations by back- propagating errors.” Nature 323 (6088): 533–536. doi:10.1038/323533a0
[20] Samuelson, P. A. (1964). “Theoretical Notes on Trade Problems,”Review of Economics and Statistics 46 , 145-154.
[21] Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. and Salakhutdinov, R. (2014). ”Dropout: a simple way to prevent neural networks from overfitting.” Journal of Ma- chine Learning Research, 15:1929–1958.
[22] Sun,S.,Wei,Y.andWang,S.(2018).”AdaBoost-LSTMEnsembleLearningforFinancialTime Series Forecasting,” Computational Science –ICCS 2018, 590-597. doi: 10.1007/978-3-319- 93713-7_55
[23] Sutskever, I., Vinyals, O. and Le, Q. V.(2014). ”Sequence to Sequence Learning with Neural Networks,” Advances in Neural Information Processing Systems 27. Curran Associates, Inc. 3104–3112.
[24] Tenti, P.(1996). ”Forecasting Foreign Exchange Rates Using Recurrent Neural Networks,” Ap- plied Artificial Intelligence, 10:6, 567-582, doi: 10.1080/088395196118434
[25] Wang, Y., Hui, X. and Soofi, A.S. (2007). ”Estimating Renminbi (RMB) Equilibrium Exchange Rate,” Journal of Policy Modeling 29(3), 417-429. doi: 10.1016/j.jpolmod.2006.12.003
[26] West, K. D. (1996). ”Asymptotic Inference About Predictive Ability,” Econometrica, Vol. 64, No. 5, 1067-1084.
[27] Zhang, B. (2018). ”Foreign Exchange Rates Forecasting with an EMD-LSTM Neural Networks Model,” Journal of Physics: Conf. Series 1053, 012005. doi: 10.1088/1742- 6596/1053/1/012005
[28] Zhang,Z.(2010).”UnderstandingtheBehavioralEquilibriumExchangeRateModelviaItsAp- plication to the Valuation of Chinese Renminbi”. doi: 10.2139/ssrn.2120419
[29] Zhuge, Q., Xu, L. and Zhang G. (2017). ”LSTM Neural Network with Emotional Analysis for Prediction of Stock Price,” Engineering Letters, 25:2, EL_25_2_09
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔